Running the Operation

Strategy defines where the organization wants to go. Operations determines whether it gets there. The work in between is practical: turning direction into plans, allocating resources, coordinating delivery, measuring performance, correcting deviations and improving the way work is done.

Operational excellence is not a state. It is the ability to run, measure, adjust and improve continuously.

What running the operation actually means

Running the operation is the management system that turns strategic intent into delivered work. It sits between strategy and day-to-day execution. Strategy defines priorities and direction. Operations translates that direction into capacity, responsibilities, workflows, projects, measures, review cycles and concrete results.

The important distinction is between working in the operation and working on it. Working in the operation means solving today’s issue, clearing the escalation, responding to the customer and keeping delivery moving. Working on the operation means asking why the issue appeared, whether the plan was realistic, which signal should have exposed the problem earlier and what has to change so the same problem does not return.

Every leader runs an operation, whether the unit contains six people or six thousand. The scale changes, but the management questions remain similar: what are we trying to deliver, how should the work run, how do we know whether it is working, what do we change when reality differs from plan and how do we make the improvement part of normal work?

Where are you now?

Operational capability tends to sit in one of three stages. These stages describe how performance is produced and how much management attention is required to keep it stable.

Stage 1: The operation runs on people

Delivery depends on individuals knowing what to do and caring enough to do it. Much of the operating knowledge sits in experienced heads. Problems are solved by whoever notices them, and recurring issues are tolerated because the immediate case gets fixed. The absence of a key person becomes visible quickly.

Stage 2: The operation runs on systems you maintain

Processes exist, measures exist and responsibilities are defined. The operation is more predictable, but it still depends heavily on management attention. Standards drift when nobody watches, improvements decay after a few months and a meaningful share of leadership time is spent restoring things that were supposed to be settled.

Stage 3: The operation improves itself

Problems surface through the operating system before they become customer complaints. Decisions are made where the information sits. Teams improve the way work is done and the improvements remain in place because they are built into the workflow, measures and routines. Leadership attention moves from daily exceptions toward capacity, constraints and future capability.

What would stop working if you were unavailable for a month?

The answer reveals where the operation still depends on personal knowledge, personal decisions or personal relationships. Those dependencies are not automatically bad, but they are useful indicators of where capability has not yet been converted into a repeatable system.

Why operations typically break down

Operational failure usually develops gradually. The patterns are visible long before they become major problems, but only if the management system is looking in the right places.

The head-knowledge trap

The operation exists largely in people’s heads. This feels efficient while experienced people are available. It becomes expensive when somebody leaves, changes role or is absent at the wrong moment. Undocumented competence can look like flexibility while creating hidden dependency.

The metric trap

The organization measures what is easy to collect rather than what reveals performance. People then optimize the number because the number has become the signal of success. Response time improves while quality falls. Utilization rises while resilience disappears. The measure begins to shape the operation more strongly than the stated objective.

The efficiency trap

Buffers are removed until the operation has no capacity to absorb normal variation. High utilization looks efficient on paper, but an operation with no slack cannot respond well to disruption, improvement work or unexpected demand. Ordinary fluctuation starts to consume management attention.

The escalation trap

Decision rights are unclear, so the safe move is upward. Each escalation makes sense individually, but together they create a bottleneck. The people closest to the information wait for approval, while senior leaders spend their time solving issues the system should have resolved lower down.

The improvement-decay trap

A better process is introduced during a project, documented and announced. Under pressure, people gradually return to the previous method because the new way requires extra effort or is not reflected in the measures and routines. Improvement that is added on top of the work tends to disappear. Improvement built into the work tends to stay.

The operating cycle

Running the operation is built from ten connected capabilities. On this page, we tell their story through five management moves: design the run, plan and deliver, measure and adjust, improve and standardize, and scale performance with AI. These moves are not a hierarchy. They form a loop. The operation runs, performance is measured, deviations are understood, the system is improved and the next cycle begins with a better design.

Design the run

Before improving an operation, you need to understand how it actually works and define how it should work. The first management task is to make the system visible, remove unnecessary dependency and decide where responsibility should sit.

Diagnosing Your Operation

Seeing what is actually happening rather than what the reporting says. Where does work wait, where is it redone, where does information disappear, which dependencies create delay and which recurring problems share the same upstream cause? Diagnosis prevents the organization from applying a familiar solution to the wrong problem.

Explore Diagnosing Your Operation

Building Systems That Work Without You

Turning individual competence into organizational capability. Clear handovers, defined decision rights, repeatable methods, accessible knowledge and deliberate removal of unnecessary leadership dependency. The test is not whether the operation works while the leader is watching. It is whether performance remains stable when that attention moves elsewhere.

Explore Building Systems That Work Without You

Plan and deliver

Once the operating model is clear, strategic direction has to become actual work. Resources, ownership, dependencies, schedules, capacity and priorities have to be translated into a plan that people can execute under real conditions.

Project and Program Management

Delivering defined outcomes alongside the continuous operation. Projects fail less often because the schedule was technically wrong than because ownership was unclear, dependencies were underestimated or the people expected to deliver the project were already committed elsewhere. Good project and program management makes the work, constraints and decisions visible early enough to act.

Explore Project and Program Management

Leading Operational People and Culture

The operating system only works when people use it, improve it and make sensible decisions when the written method no longer fits the situation. Operational culture is visible in what happens under pressure. If the real rules differ from the documented rules, people will follow the real ones.

Explore Leading Operational People and Culture

Measure and adjust

Planning creates an expectation. Measurement tests that expectation against reality. The purpose of operational control is not to report what happened. It is to identify the difference between expected and actual performance early enough to decide what needs to change.

Performance Measurement and Control

Defining what good looks like, choosing indicators that reveal whether the operation is moving toward it and understanding what behavior those indicators create. Good control is a loop: expected performance, actual performance, gap, cause, adjustment and measurement again. The best measures make developing problems visible before the customer does.

Explore Performance Measurement and Control

Measure reality against the plan, understand the difference, adjust what matters and run the cycle again.

Improve and standardize

Once the difference between expected and actual performance is understood, the operation can improve deliberately. Improvement becomes valuable when the better method is embedded into normal work rather than remaining a project recommendation.

Strategic Operations and Transformation

Changing how work is designed when incremental improvement is no longer enough. Which activities should be simplified, combined, removed, moved, sourced differently or supported by new capability? Transformation becomes operational when the redesigned method is reflected in roles, workflows, technology, measures and management routines.

Explore Strategic Operations and Transformation

Standardization does not mean freezing the operation. It means turning a proven improvement into the new starting point. The next cycle can then challenge that standard again when evidence shows that a better way is possible.

Scale performance with AI

AI changes operations most when it becomes part of the workflow rather than an isolated tool. The near-term value starts with reporting, planning, document handling and repetitive administrative work. The larger shift comes when AI-supported and agentic workflows coordinate tasks, monitor conditions, trigger actions and improve operational performance continuously.

Operational AI Basics: Quick Wins in Reporting, Planning, Admin

Practical applications that reduce administrative load and management effort without redesigning the entire operation. Reporting, scheduling, document handling, first-line triage and routine preparation can create visible value quickly when the underlying process is already understood.

Explore Operational AI Basics

AI and Automation in Operations

Using AI and automation to improve throughput, quality, cost, speed and consistency. The important question is not what can be automated, but which workflow should exist before automation begins. A weak process automated at scale remains a weak process, only faster and harder to inspect.

Explore AI and Automation in Operations

Autonomous and Self-Optimizing Operations

Agentic and autonomous workflows that monitor conditions, coordinate work, adjust plans and trigger actions with decreasing human intervention. The management challenge shifts toward objectives, boundaries, exception handling, accountability and the ability to intervene when the automated system optimizes the wrong thing.

Explore Autonomous and Self-Optimizing Operations

Resilient Operations: Preparing for Disruption

Designing for the case where normal operation is no longer possible. Which dependencies can fail, where are the single points of failure, what does degraded operation look like and which decisions must remain available when automation, suppliers, infrastructure or information are disrupted? Resilience keeps performance recoverable when efficiency alone would make the system brittle.

Explore Resilient Operations

How a good operation looks like

A well-run operation is quiet. Work moves without constant chasing. Problems appear in the measures before they appear in complaints. Handovers happen because the workflow requires them, not because somebody remembered to follow up. New people become productive faster because there is a defined system to learn.

• The plan is visible. People know what has to be delivered, by when, with which resources and dependencies.
• Performance is measurable. The operation knows what good looks like and detects meaningful deviation early.
• Adjustment is normal. Plans change when reality changes, without turning every correction into a crisis.
• Problems improve the system. Recurring issues trigger root-cause work rather than repeated firefighting.
• Improvements become the new standard. Better methods remain in place after the project or workshop has ended.
• AI improves operational results. Automation and agentic workflows are judged by measurable impact on performance, not by technical novelty.
• The operation remains recoverable. Critical dependencies, exceptions and degraded modes are understood before disruption occurs.

The objective is not to remove management from the operation. It is to move management attention from repeated intervention toward design, performance, improvement and future capability.

How Running the Operation connects to the rest of the framework

Thinking Strategically defines direction and priorities. Running the Operation turns that direction into plans, capacity, work and measurable results. Better Decisions determines how choices are made when the operating data shows that reality differs from the plan.

Leading People shapes whether the operation depends on heroics or develops capability across the team. Communicating with Impact ensures that priorities, standards and changes are understood consistently. Facilitating and Collaborating determines whether reviews, projects and improvement work produce decisions rather than discussion.

Leading Change becomes essential when operational improvement requires new behavior, new roles or a new operating model. AI for Executives increasingly affects the entire operating cycle, from planning and reporting to agentic workflows and autonomous execution. Timeless Leadership Wisdom adds an older perspective on delegation, logistics, standardization, discipline and the practical consequences of weak operational design.

Frequently asked questions about running the operation

What is the difference between strategy and operations?

Strategy defines direction, priorities and major choices. Operations translates those choices into plans, resources, workflows, responsibilities, measures and delivery. Strategy answers where the organization intends to go. Operations determines whether the organization can get there repeatedly.

What is operational excellence?

Operational excellence is not a fixed end state. It is the capability to run the operation reliably, measure performance, adjust when reality differs from plan and continuously improve the way work is done.

How do I know which KPIs to use?

Start with the result the operation is expected to deliver and ask which indicators would show meaningful deviation early. Then test the behavior each measure is likely to create. A useful KPI should help management understand performance and decide what to adjust, not simply make reporting easier.

Why do process improvements fail to stick?

Usually because the improved method remains an additional requirement rather than becoming the normal way of working. If it is not reflected in the workflow, measurement, tools and management routines, pressure will eventually pull the operation back toward the previous method.

Where should I start with AI in operations?

Start with a process you understand and an operational problem you can measure. Administrative work such as reporting, scheduling, document handling and triage is often a practical starting point. Use the result to learn how AI changes the workflow before moving into more autonomous or higher-risk operating processes.

What will agentic workflows change in operations?

Agentic workflows can increasingly coordinate tasks, monitor conditions, trigger actions and adjust plans across multiple steps without waiting for manual instruction each time. The management focus shifts toward objectives, boundaries, performance measures, exceptions and accountability for the overall result.

How do I improve an operation when there is no time to improve it?

Lack of improvement capacity is itself an operational signal. An operation running at full utilization has little room to change. The first improvement may therefore be to create capacity by stopping low-value work, simplifying reporting, reducing avoidable escalation or accepting temporary reduction in a non-critical area.

Where to start

Choose one important operational outcome and write down what good performance for running the operation should look like. Then compare that expectation with what is happening now. Identify the most meaningful gap, understand why it exists and decide which part of the operating system should change before the next cycle.

1. Define the expected result. What should the operation deliver and under which conditions?
2. Measure reality. What is actually happening, and where does it differ from the expectation?
3. Adjust the system. What needs to change in plan, workflow, responsibility, capacity or technology?
4. Standardize the improvement. How does the better method become the normal method?
5. Run the cycle again. Measure the result, learn from the difference and improve the next version.

Running the operation means turning strategic direction into a management cycle that keeps learning from reality.

Explore the 10 Operational Capabilities